« Previous | Contents | Next »
Listen
Review of Marches and Parades: A Survey of Views Across Scotland
ANNEX 1 SAMPLE DESIGN AND WEIGHTING
The sample was based on a Random Digit Dial (RDD) survey with quotas attached. RDD works on the basis of a sample of listed telephone numbers, with the last digit or digits being changed at random to ensure coverage of unlisted (ex-directory) numbers. A random sample of telephone numbers was drawn within each police force area. This was feasible because police force area boundaries are contiguous with local authority boundaries. 12 Although area dialling codes are not precisely contiguous with local authority boundaries, in the vast majority of cases it is possible to identify the local authority area from the dialling code. As a further check on accuracy, we included a check on local authority area within the questionnaire.
The aim was to undertake a total of 675 interviews: 150 interviews in the Strathclyde Police area, and a further 75 interviews in each of the other seven police force areas. In the event, 138 interviews were achieved in Strathclyde and between 69 and 87 in each of the other 7 police force areas (see Table A1.1 below). This sample size allowed analysis of differences by police force area and, at a national level, by key demographic variables (age, sex, social class, and religious affiliation).
Quotas were set on age, sex and working status to ensure a sample that was broadly representative in terms of these key demographic variables. These quotas were based on the population profile within each police force area.
The data was weighted to take account of the disproportionate sampling by police force area. For example, respondents from Strathclyde represented 20.4% of the total sample although they account for 43.6% of the Scottish population. We therefore weighted the Strathclyde data by a factor of 2.14 13. The weightings which were applied are shown in Table A1.1 below. No other weightings (e.g. on demographic variables) were applied.
Table A1.1 Sample and weighting by police force area
Police Force Area | Population | % of Scottish population | Achieved interviews (unweighted base) | Weighting applied | % of sample after weighting | Weighted base |
Central Scotland | 279,000 | 5.5 | 78 | 0.48 | 5.5 | 37 |
Dumfries & Galloway | 148,000 | 2.9 | 72 | 0.27 | 3.0 | 20 |
Fife | 349,000 | 6.9 | 75 | 0.62 | 7.0 | 47 |
Grampian | 526,000 | 10.4 | 73 | 0.96 | 10.4 | 70 |
Lothian & Borders | 886,000 | 17.5 | 87 | 1.36 | 17.5 | 118 |
Northern | 277,000 | 5.5 | 69 | 0.54 | 5.5 | 37 |
Strathclyde | 2,207,000 | 43.6 | 138 | 2.14 | 43.6 | 295 |
Tayside | 389,000 | 7.7 | 84 | 0.62 | 7.7 | 52 |
Total | 5,061,000 | 100 | 676 | | 100 | 676 |
The demographic profile of the achieved sample is shown in Table A1.2.
Table A1.2 Sample profile
| Base (unweighted) | % (unweighted) | Base (weighted) | %
(weighted) |
Male | 286 | 42% | 282 | 42% |
Female | 390 | 58% | 394 | 58% |
Aged 16-29 years | 101 | 15% | 102 | 15% |
Aged 30-44 years | 190 | 28% | 193 | 29% |
Aged 45-59 years | 190 | 28% | 181 | 27% |
Aged 60+ years | 195 | 29% | 200 | 30% |
Social class ABC1 | 338 | 50% | 335 | 50% |
Social class C2DE | 338 | 50% | 340 | 50% |
Working | 366 | 54% | 350 | 52% |
Not working | 307 | 45% | 323 | 48% |
Brought up as Protestant | 456 | 67% | 433 | 64% |
Brought up as Catholic | 96 | 14% | 125 | 18% |
Brought up in other/no religion | 115 | 17% | 107 | 16% |
Categories may not always sum to 676 or 100% due to non-response and/or rounding
« Previous | Contents | Next »